@@ -404,46 +400,38 @@ class Markdown_Parser {
# the inner nested divs must be indented.
# We need to do this before the next, more liberal match, because the next
# match will start at the first `
` and stop at the first `
`.
- $text = preg_replace_callback('{
- ( # save in $1
- ^ # start of line (with /m)
- <('.$block_tags_a.')# start tag = $2
- '.$attr.'>\n # attributes followed by > and \n
- '.$content.' # content, support nesting
- # the matching end tag
- [ ]* # trailing spaces/tabs
- (?=\n+|\Z) # followed by a newline or end of document
- )
- }xmi',
- array(&$this, '_hashHTMLBlocks_callback'),
- $text);
+ $text = preg_replace_callback('{(?>
+ (?>
+ (?<=\n\n) # Starting after a blank line
+ | # or
+ \A\n? # the beginning of the doc
+ )
+ ( # save in $1
- #
- # Match from `\n
` to `\n`, handling nested tags in between.
- #
- $text = preg_replace_callback('{
- ( # save in $1
- ^ # start of line (with /m)
+ # Match from `\n
` to `\n`, handling nested tags
+ # in between.
+
+ [ ]{0,'.$less_than_tab.'}
<('.$block_tags_b.')# start tag = $2
- '.$attr.'> # attributes followed by >
+ '.$attr.'> # attributes followed by > and \n
'.$content.' # content, support nesting
# the matching end tag
[ ]* # trailing spaces/tabs
(?=\n+|\Z) # followed by a newline or end of document
- )
- }xmi',
- array(&$this, '_hashHTMLBlocks_callback'),
- $text);
- # Special case just for
. It was easier to make a special case than
- # to make the other regex more complicated.
- $text = preg_replace_callback('{
- (?:
- (?<=\n\n) # Starting after a blank line
- | # or
- \A\n? # the beginning of the doc
- )
- ( # save in $1
+ | # Special version for tags of group a.
+
+ [ ]{0,'.$less_than_tab.'}
+ <('.$block_tags_a.')# start tag = $3
+ '.$attr.'>[ ]*\n # attributes followed by >
+ '.$content2.' # content, support nesting
+ # the matching end tag
+ [ ]* # trailing spaces/tabs
+ (?=\n+|\Z) # followed by a newline or end of document
+
+ | # Special case just for
. It was easier to make a special
+ # case than to make the other regex more complicated.
+
[ ]{0,'.$less_than_tab.'}
<(hr) # start tag = $2
\b # word break
@@ -451,38 +439,18 @@ class Markdown_Parser {
/?> # the matching end tag
[ ]*
(?=\n{2,}|\Z) # followed by a blank line or end of document
- )
- }xi',
- array(&$this, '_hashHTMLBlocks_callback'),
- $text);
-
- # Special case for standalone HTML comments:
- $text = preg_replace_callback('{
- (?:
- (?<=\n\n) # Starting after a blank line
- | # or
- \A\n? # the beginning of the doc
- )
- ( # save in $1
+
+ | # Special case for standalone HTML comments:
+
[ ]{0,'.$less_than_tab.'}
(?s:
)
[ ]*
(?=\n{2,}|\Z) # followed by a blank line or end of document
- )
- }x',
- array(&$this, '_hashHTMLBlocks_callback'),
- $text);
-
- # PHP and ASP-style processor instructions (hashBlock($text);
return "\n\n$key\n\n";
}
-
-
- function hashBlock($text) {
+
+
+ function hashPart($text, $boundary = 'X') {
#
- # Called whenever a tag must be hashed when a function insert a block-level
- # tag in $text, it pass through this function and is automaticaly escaped,
- # which remove the need to call _HashHTMLBlocks at every step.
+ # Called whenever a tag must be hashed when a function insert an atomic
+ # element in the text stream. Passing $text to through this function gives
+ # a unique text-token which will be reverted back when calling unhash.
+ #
+ # The $boundary argument specify what character should be used to surround
+ # the token. By convension, "B" is used for block elements that needs not
+ # to be wrapped into paragraph tags at the end, ":" is used for elements
+ # that are word separators and "S" is used for general span-level elements.
#
# Swap back any tag hash found in $text so we do not have to `unhash`
# multiple times at the end.
$text = $this->unhash($text);
# Then hash the block.
- $key = "B\x1A". md5($text);
+ static $i = 0;
+ $key = "$boundary\x1A" . ++$i . $boundary;
$this->html_hashes[$key] = $text;
- $this->html_blocks[$key] = $text;
return $key; # String that will replace the tag.
}
- function hashSpan($text, $word_separator = false) {
+ function hashBlock($text) {
#
- # Called whenever a tag must be hashed when a function insert a span-level
- # element in $text, it pass through this function and is automaticaly
- # escaped, blocking invalid nested overlap. If optional argument
- # $word_separator is true, surround the hash value by spaces.
+ # Shortcut function for hashPart with block-level boundaries.
#
- # Swap back any tag hash found in $text so we do not have to `unhash`
- # multiple times at the end.
- $text = $this->unhash($text);
-
- # Then hash the span.
- $key = "S\x1A". md5($text);
- if ($word_separator) $key = ":$key:";
-
- $this->html_hashes[$key] = $text;
- return $key; # String that will replace the span tag.
+ return $this->hashPart($text, 'B');
}
@@ -590,9 +552,16 @@ class Markdown_Parser {
function doHorizontalRules($text) {
# Do Horizontal Rules:
return preg_replace(
- array('{^[ ]{0,2}([ ]?\*[ ]?){3,}[ ]*$}mx',
- '{^[ ]{0,2}([ ]? -[ ]?){3,}[ ]*$}mx',
- '{^[ ]{0,2}([ ]? _[ ]?){3,}[ ]*$}mx'),
+ '{
+ ^[ ]{0,3} # Leading space
+ ([*-_]) # $1: First marker
+ (?> # Repeated marker group
+ [ ]{0,2} # Zero, one, or two spaces.
+ \1 # Marker character
+ ){2,} # Group repeated at least twice
+ [ ]* # Tailing spaces
+ $ # End of line.
+ }mx',
"\n".$this->hashBlock("
empty_element_suffix")."\n",
$text);
}
@@ -603,9 +572,9 @@ class Markdown_Parser {
# These are all the transformations that occur *within* block-level
# tags like paragraphs, headers, and list items.
#
- "escapeSpecialCharsWithinTagAttributes" => -20,
- "doCodeSpans" => -10,
- "encodeBackslashEscapes" => -5,
+ # Process character escapes, code spans, and inline HTML
+ # in one shot.
+ "parseSpan" => -30,
# Process anchor and image tags. Images must come first,
# because ![foo][f] looks like an anchor.
@@ -636,35 +605,11 @@ class Markdown_Parser {
function doHardBreaks($text) {
# Do hard breaks:
- $br_tag = $this->hashSpan("empty_element_suffix\n");
- return preg_replace('/ {2,}\n/', $br_tag, $text);
+ return preg_replace_callback('/ {2,}\n/',
+ array(&$this, '_doHardBreaks_callback'), $text);
}
-
-
- function escapeSpecialCharsWithinTagAttributes($text) {
- #
- # Within tags -- meaning between < and > -- encode [\ ` * _] so they
- # don't conflict with their use in Markdown for code, italics and strong.
- # We're replacing each such character with its corresponding MD5 checksum
- # value; this is likely overkill, but it should prevent us from colliding
- # with the escape values by accident.
- #
- if ($this->no_markup) return $text;
-
- $tokens = $this->tokenizeHTML($text);
- $text = ''; # rebuild $text from the tokens
-
- foreach ($tokens as $cur_token) {
- if ($cur_token[0] == 'tag') {
-// $cur_token[1] = str_replace('\\', $this->escape_table['\\'], $cur_token[1]);
-// $cur_token[1] = str_replace('`', $this->escape_table['`'], $cur_token[1]);
-// $cur_token[1] = str_replace('*', $this->escape_table['*'], $cur_token[1]);
-// $cur_token[1] = str_replace('_', $this->escape_table['_'], $cur_token[1]);
- $cur_token[1] = $this->hashSpan($cur_token[1]);
- }
- $text .= $cur_token[1];
- }
- return $text;
+ function _doHardBreaks_callback($matches) {
+ return $this->hashPart("empty_element_suffix\n");
}
@@ -765,7 +710,7 @@ class Markdown_Parser {
$link_text = $this->runSpanGamut($link_text);
$result .= ">$link_text";
- $result = $this->hashSpan($result);
+ $result = $this->hashPart($result);
}
else {
$result = $whole_match;
@@ -790,7 +735,7 @@ class Markdown_Parser {
$link_text = $this->runSpanGamut($link_text);
$result .= ">$link_text";
- return $this->hashSpan($result);
+ return $this->hashPart($result);
}
@@ -867,7 +812,7 @@ class Markdown_Parser {
$result .= " title=\"$title\"";
}
$result .= $this->empty_element_suffix;
- $result = $this->hashSpan($result);
+ $result = $this->hashPart($result);
}
else {
# If there's no such link ID, leave intact:
@@ -890,7 +835,7 @@ class Markdown_Parser {
}
$result .= $this->empty_element_suffix;
- return $this->hashSpan($result);
+ return $this->hashPart($result);
}
@@ -902,10 +847,8 @@ class Markdown_Parser {
# Header 2
# --------
#
- $text = preg_replace_callback('{ ^(.+?)[ ]*\n=+[ ]*\n+ }mx',
- array(&$this, '_doHeaders_callback_setext_h1'), $text);
- $text = preg_replace_callback('{ ^(.+?)[ ]*\n-+[ ]*\n+ }mx',
- array(&$this, '_doHeaders_callback_setext_h2'), $text);
+ $text = preg_replace_callback('{ ^(.+?)[ ]*\n(=+|-+)[ ]*\n+ }mx',
+ array(&$this, '_doHeaders_callback_setext'), $text);
# atx-style headers:
# # Header 1
@@ -926,12 +869,9 @@ class Markdown_Parser {
return $text;
}
- function _doHeaders_callback_setext_h1($matches) {
- $block = "".$this->runSpanGamut($matches[1])."
";
- return "\n" . $this->hashBlock($block) . "\n\n";
- }
- function _doHeaders_callback_setext_h2($matches) {
- $block = "".$this->runSpanGamut($matches[1])."
";
+ function _doHeaders_callback_setext($matches) {
+ $level = $matches[2]{0} == '=' ? 1 : 2;
+ $block = "".$this->runSpanGamut($matches[1])."";
return "\n" . $this->hashBlock($block) . "\n\n";
}
function _doHeaders_callback_atx($matches) {
@@ -1091,8 +1031,8 @@ class Markdown_Parser {
$text = preg_replace_callback('{
(?:\n\n|\A)
( # $1 = the code block -- one or more lines, starting with a space/tab
- (?:
- (?:[ ]{'.$this->tab_width.'} | \t) # Lines must start with a tab or a tab-width of spaces
+ (?>
+ [ ]{'.$this->tab_width.'} # Lines must start with a tab or a tab-width of spaces
.*\n+
)+
)
@@ -1105,82 +1045,23 @@ class Markdown_Parser {
function _doCodeBlocks_callback($matches) {
$codeblock = $matches[1];
- $codeblock = $this->encodeCode($this->outdent($codeblock));
-// $codeblock = $this->detab($codeblock);
- # trim leading newlines and trailing whitespace
+ $codeblock = $this->outdent($codeblock);
+ $codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES);
+
+ # trim leading newlines and trailing newlines
$codeblock = preg_replace(array('/\A\n+/', '/\n+\z/'), '', $codeblock);
- $result = "\n\n".$this->hashBlock("" . $codeblock . "\n
")."\n\n";
-
- return $result;
+ $codeblock = "$codeblock\n
";
+ return "\n\n".$this->hashBlock($codeblock)."\n\n";
}
- function doCodeSpans($text) {
+ function makeCodeSpan($code) {
#
- # * Backtick quotes are used for Just type foo `bar` baz at the prompt.
- #
- # There's no arbitrary limit to the number of backticks you
- # can use as delimters. If you need three consecutive backticks
- # in your code, use four for delimiters, etc.
- #
- # * You can use spaces to get literal backticks at the edges:
- #
- # ... type `` `bar` `` ...
- #
- # Turns to:
- #
- # ... type `bar` ...
- #
- $text = preg_replace_callback('@
- (?encodeCode($c);
- return $this->hashSpan("$c");
- }
-
-
- function encodeCode($_) {
- #
- # Encode/escape certain characters inside Markdown code runs.
- # The point is that in code, these characters are literals,
- # and lose their special Markdown meanings.
- #
- # Encode all ampersands; HTML entities are not
- # entities within a Markdown code span.
- $_ = str_replace('&', '&', $_);
-
- # Do the angle bracket song and dance:
- $_ = str_replace(array('<', '>'),
- array('<', '>'), $_);
-
- # Now, escape characters that are magic in Markdown:
-// $_ = str_replace(array_keys($this->escape_table),
-// array_values($this->escape_table), $_);
-
- return $_;
+ $code = htmlspecialchars(trim($code), ENT_NOQUOTES);
+ return $this->hashPart("$code");
}
@@ -1217,19 +1098,19 @@ class Markdown_Parser {
function _doItalicAndBold_em_callback($matches) {
$text = $matches[2];
$text = $this->runSpanGamut($text);
- return $this->hashSpan("$text");
+ return $this->hashPart("$text");
}
function _doItalicAndBold_strong_callback($matches) {
$text = $matches[2];
$text = $this->runSpanGamut($text);
- return $this->hashSpan("$text");
+ return $this->hashPart("$text");
}
function doBlockQuotes($text) {
$text = preg_replace_callback('/
( # Wrap whole match in $1
- (
+ (?>
^[ ]*>[ ]? # ">" at the start of a line
.+\n # rest of the first line
(.+\n)* # subsequent consecutive lines
@@ -1273,24 +1154,21 @@ class Markdown_Parser {
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY);
#
- # Wrap tags.
+ # Wrap
tags and unhashify HTML blocks
#
foreach ($grafs as $key => $value) {
- if (!isset( $this->html_blocks[$value] )) {
+ if (!preg_match('/^B\x1A[0-9]+B$/', $value)) {
+ # Is a paragraph.
$value = $this->runSpanGamut($value);
$value = preg_replace('/^([ ]*)/', "
", $value);
$value .= "
";
$grafs[$key] = $this->unhash($value);
}
- }
-
- #
- # Unhashify HTML blocks
- #
- foreach ($grafs as $key => $graf) {
- # Modify elements of @grafs in-place...
- if (isset($this->html_blocks[$graf])) {
- $block = $this->html_blocks[$graf];
+ else {
+ # Is a block.
+ # Modify elements of @grafs in-place...
+ $graf = $value;
+ $block = $this->html_hashes[$graf];
$graf = $block;
// if (preg_match('{
// \A
@@ -1355,20 +1233,6 @@ class Markdown_Parser {
}
- function encodeBackslashEscapes($text) {
- #
- # Parameter: String.
- # Returns: The string, with after processing the following backslash
- # escape sequences.
- #
- # Must process escaped backslashes first (should be first in list).
- foreach ($this->backslash_escape_table as $search => $replacement) {
- $text = str_replace($search, $this->hashSpan($replacement), $text);
- }
- return $text;
- }
-
-
function doAutoLinks($text) {
$text = preg_replace_callback('{<((https?|ftp|dict):[^\'">\s]+)>}',
array(&$this, '_doAutoLinks_url_callback'), $text);
@@ -1391,12 +1255,12 @@ class Markdown_Parser {
function _doAutoLinks_url_callback($matches) {
$url = $this->encodeAmpsAndAngles($matches[1]);
$link = "$url";
- return $this->hashSpan($link);
+ return $this->hashPart($link);
}
function _doAutoLinks_email_callback($matches) {
$address = $matches[1];
$link = $this->encodeEmailAddress($address);
- return $this->hashSpan($link);
+ return $this->hashPart($link);
}
@@ -1441,73 +1305,83 @@ class Markdown_Parser {
}
- function tokenizeHTML($str) {
+ function parseSpan($str) {
#
- # Parameter: String containing HTML + Markdown markup.
- # Returns: An array of the tokens comprising the input
- # string. Each token is either a tag or a run of text
- # between tags. Each element of the array is a
- # two-element array; the first is either 'tag' or 'text';
- # the second is the actual value.
- # Note: Markdown code spans are taken into account: no tag token is
- # generated within a code span.
+ # Take the string $str and parse it into tokens, hashing embeded HTML,
+ # escaped characters and handling code spans.
#
- $tokens = array();
-
- while ($str != "") {
- #
- # Each loop iteration seach for either the next tag or the next
- # openning code span marker. If a code span marker is found, the
- # code span is extracted in entierty and will result in an extra
- # text token.
- #
- $parts = preg_split('{
+ $output = '';
+
+ $regex = '{
(
+ \\\\['.preg_quote($this->escape_chars).']
+ |
(?no_markup ? '' : '
|
# comment
|
<\?.*?\?> | <%.*?%> # processing instruction
|
<[/!$]?[-a-zA-Z0-9:]+ # regular tags
- (?:
+ (?>
\s
(?>[^"\'>]+|"[^"]*"|\'[^\']*\')*
)?
>
+ ').'
)
- }xs', $str, 2, PREG_SPLIT_DELIM_CAPTURE);
+ }xs';
+
+ while (1) {
+ #
+ # Each loop iteration seach for either the next tag, the next
+ # openning code span marker, or the next escaped character.
+ # Each token is then passed to handleSpanToken.
+ #
+ $parts = preg_split($regex, $str, 2, PREG_SPLIT_DELIM_CAPTURE);
# Create token from text preceding tag.
if ($parts[0] != "") {
- $tokens[] = array('text', $parts[0]);
+ $output .= $parts[0];
}
# Check if we reach the end.
- if (count($parts) < 3) {
- break;
+ if (isset($parts[1])) {
+ $output .= $this->handleSpanToken($parts[1], $parts[2]);
+ $str = $parts[2];
}
-
- # Create token from tag or code span.
- if ($parts[1]{0} == "`") {
- $tokens[] = array('text', $parts[1]);
- $str = $parts[2];
-
- # Skip the whole code span, pass as text token.
- if (preg_match('/^(.*(?hashPart("&#". ord($token{1}). ";");
+ case "`":
+ # Search for end marker in remaining text.
+ if (preg_match('/^(.*?[^`])'.$token.'(?!`)(.*)$/sm',
+ $str, $matches))
+ {
+ $str = $matches[2];
+ $codespan = $this->makeCodeSpan($matches[1]);
+ return $this->hashPart($codespan);
+ }
+ return $token; // return as text since no ending marker found.
+ default:
+ return $this->hashPart($token);
+ }
}
@@ -1515,7 +1389,7 @@ class Markdown_Parser {
#
# Remove one level of line-leading tabs or spaces
#
- return preg_replace("/^(\\t|[ ]{1,$this->tab_width})/m", "", $text);
+ return preg_replace('/^(\t|[ ]{1,'.$this->tab_width.'})/m', '', $text);
}
@@ -1531,26 +1405,46 @@ class Markdown_Parser {
# tab characters. Then we reconstruct every line by adding the
# appropriate number of space between each blocks.
- $strlen = $this->utf8_strlen; # strlen function for UTF-8.
- $lines = explode("\n", $text);
- $text = "";
+ $text = preg_replace_callback('/^.*\t.*$/m',
+ array(&$this, '_detab_callback'), $text);
- foreach ($lines as $line) {
- # Split in blocks.
- $blocks = explode("\t", $line);
- # Add each blocks to the line.
- $line = $blocks[0];
- unset($blocks[0]); # Do not add first block twice.
- foreach ($blocks as $block) {
- # Calculate amount of space, insert spaces, insert block.
- $amount = $this->tab_width -
- $strlen($line, 'UTF-8') % $this->tab_width;
- $line .= str_repeat(" ", $amount) . $block;
- }
- $text .= "$line\n";
- }
+// $strlen = $this->utf8_strlen; # strlen function for UTF-8.
+// $lines = explode("\n", $text);
+// $text = "";
+//
+// foreach ($lines as $line) {
+// # Split in blocks.
+// $blocks = explode("\t", $line);
+// # Add each blocks to the line.
+// $line = $blocks[0];
+// unset($blocks[0]); # Do not add first block twice.
+// foreach ($blocks as $block) {
+// # Calculate amount of space, insert spaces, insert block.
+// $amount = $this->tab_width -
+// $strlen($line, 'UTF-8') % $this->tab_width;
+// $line .= str_repeat(" ", $amount) . $block;
+// }
+// $text .= "$line\n";
+// }
return $text;
}
+ function _detab_callback($matches) {
+ $line = $matches[0];
+ $strlen = $this->utf8_strlen; # strlen function for UTF-8.
+
+ # Split in blocks.
+ $blocks = explode("\t", $line);
+ # Add each blocks to the line.
+ $line = $blocks[0];
+ unset($blocks[0]); # Do not add first block twice.
+ foreach ($blocks as $block) {
+ # Calculate amount of space, insert spaces, insert block.
+ $amount = $this->tab_width -
+ $strlen($line, 'UTF-8') % $this->tab_width;
+ $line .= str_repeat(" ", $amount) . $block;
+ }
+ return $line;
+ }
function _initDetab() {
#
# Check for the availability of the function in the `utf8_strlen` property
@@ -1569,13 +1463,15 @@ class Markdown_Parser {
#
# Swap back in all the tags hashed by _HashHTMLBlocks.
#
- return str_replace(array_keys($this->html_hashes),
- array_values($this->html_hashes), $text);
+ return preg_replace_callback('/(.)\x1A[0-9]+\1/',
+ array(&$this, '_unhash_callback'), $text);
+ }
+ function _unhash_callback($matches) {
+ return $this->html_hashes[$matches[0]];
}
}
-
/*
PHP Markdown
@@ -1618,6 +1514,8 @@ Version History
See the readme file for detailed release notes for this version.
+1.0.1i (28 Aug 2007)
+
1.0.1h (3 Aug 2007)
1.0.1g (3 Jul 2007)